Prediction

ID

i5xhgehjkjfyhlx7dg2f5tnrti

Status

Succeeded

Source

Web

Hardware

–

Total duration

1m 41s

Created

over 3 years ago by @chenxwh

Input

task: image_captioning
image

{
  "task": "image_captioning",
  "image": "https://replicate.delivery/mgxm/f4e50a7b-e8ca-432f-8e68-082034ebcc70/demo.jpg"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "image_captioning",
      image: "https://replicate.delivery/mgxm/f4e50a7b-e8ca-432f-8e68-082034ebcc70/demo.jpg"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "image_captioning",
        "image": "https://replicate.delivery/mgxm/f4e50a7b-e8ca-432f-8e68-082034ebcc70/demo.jpg"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "image_captioning",
      "image": "https://replicate.delivery/mgxm/f4e50a7b-e8ca-432f-8e68-082034ebcc70/demo.jpg"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="image_captioning"' \
  -i 'image="https://replicate.delivery/mgxm/f4e50a7b-e8ca-432f-8e68-082034ebcc70/demo.jpg"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "image_captioning",
      "image": "https://replicate.delivery/mgxm/f4e50a7b-e8ca-432f-8e68-082034ebcc70/demo.jpg"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Caption: a woman sitting on the beach with a dog

Generated in

6.7 seconds

salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099

Prediction

Model

ID

m2ps5uu7rbbnxpfwmdebf7qnr4

Status

Succeeded

Source

Web

Hardware

–

Total duration

1.0s

Created

over 3 years ago by @chenxwh

Input

task: image_text_matching
image
caption: a dog and a women are sitting at the beach

{
  "task": "image_text_matching",
  "image": "https://replicate.delivery/mgxm/222a1a8a-9ebe-4d38-82cb-9d755020efc7/demo.jpg",
  "caption": "a dog and a women are sitting at the beach"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "image_text_matching",
      image: "https://replicate.delivery/mgxm/222a1a8a-9ebe-4d38-82cb-9d755020efc7/demo.jpg",
      caption: "a dog and a women are sitting at the beach"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "image_text_matching",
        "image": "https://replicate.delivery/mgxm/222a1a8a-9ebe-4d38-82cb-9d755020efc7/demo.jpg",
        "caption": "a dog and a women are sitting at the beach"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "image_text_matching",
      "image": "https://replicate.delivery/mgxm/222a1a8a-9ebe-4d38-82cb-9d755020efc7/demo.jpg",
      "caption": "a dog and a women are sitting at the beach"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="image_text_matching"' \
  -i 'image="https://replicate.delivery/mgxm/222a1a8a-9ebe-4d38-82cb-9d755020efc7/demo.jpg"' \
  -i 'caption="a dog and a women are sitting at the beach"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "image_text_matching",
      "image": "https://replicate.delivery/mgxm/222a1a8a-9ebe-4d38-82cb-9d755020efc7/demo.jpg",
      "caption": "a dog and a women are sitting at the beach"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

The image and text is matched with a probability of 0.9977. The image feature and text feature has a cosine similarity of 0.5177.

Generated in

0.8 seconds

salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099

Prediction

Model

ID

fmljgux3ajgn5knswprvff4tkq

Status

Succeeded

Source

Web

Hardware

–

Total duration

0.9s

Created

over 3 years ago by @chenxwh

Input

task: image_text_matching
image
caption: a dog and a cat are playing in the garden

{
  "task": "image_text_matching",
  "image": "https://replicate.delivery/mgxm/2c0e8a0d-232d-4cfa-b97a-b3258be2a2e5/demo.jpg",
  "caption": "a dog and a cat are playing in the garden"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "image_text_matching",
      image: "https://replicate.delivery/mgxm/2c0e8a0d-232d-4cfa-b97a-b3258be2a2e5/demo.jpg",
      caption: "a dog and a cat are playing in the garden"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "image_text_matching",
        "image": "https://replicate.delivery/mgxm/2c0e8a0d-232d-4cfa-b97a-b3258be2a2e5/demo.jpg",
        "caption": "a dog and a cat are playing in the garden"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "image_text_matching",
      "image": "https://replicate.delivery/mgxm/2c0e8a0d-232d-4cfa-b97a-b3258be2a2e5/demo.jpg",
      "caption": "a dog and a cat are playing in the garden"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="image_text_matching"' \
  -i 'image="https://replicate.delivery/mgxm/2c0e8a0d-232d-4cfa-b97a-b3258be2a2e5/demo.jpg"' \
  -i 'caption="a dog and a cat are playing in the garden"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "image_text_matching",
      "image": "https://replicate.delivery/mgxm/2c0e8a0d-232d-4cfa-b97a-b3258be2a2e5/demo.jpg",
      "caption": "a dog and a cat are playing in the garden"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

The image and text is matched with a probability of 0.0000. The image feature and text feature has a cosine similarity of 0.2619.

Generated in

0.7 seconds

salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099

Prediction

Model

ID

kt5yvsmuhfb7bfc5gnkzor2i3y

Status

Succeeded

Source

Web

Hardware

–

Total duration

2m 36s

Created

over 3 years ago by @lijunnan1992

Input

task: visual_question_answering
image
question: where is the woman?

{
  "task": "visual_question_answering",
  "image": "https://replicate.delivery/mgxm/32e7126a-dd86-4e8c-8706-5855b8b69cf1/demo.jpg",
  "question": "where is the woman?"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "visual_question_answering",
      image: "https://replicate.delivery/mgxm/32e7126a-dd86-4e8c-8706-5855b8b69cf1/demo.jpg",
      question: "where is the woman?"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "visual_question_answering",
        "image": "https://replicate.delivery/mgxm/32e7126a-dd86-4e8c-8706-5855b8b69cf1/demo.jpg",
        "question": "where is the woman?"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/32e7126a-dd86-4e8c-8706-5855b8b69cf1/demo.jpg",
      "question": "where is the woman?"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="visual_question_answering"' \
  -i 'image="https://replicate.delivery/mgxm/32e7126a-dd86-4e8c-8706-5855b8b69cf1/demo.jpg"' \
  -i 'question="where is the woman?"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/32e7126a-dd86-4e8c-8706-5855b8b69cf1/demo.jpg",
      "question": "where is the woman?"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Answer: on beach

Generated in

9.1 seconds

salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099

Prediction

Model

ID

wwm6t4xre5al3mnman2v5hymfm

Status

Succeeded

Source

Web

Hardware

–

Total duration

4m 21s

Created

over 3 years ago

Input

task: image_captioning
image

{
  "task": "image_captioning",
  "image": "https://replicate.delivery/mgxm/13f7ddbb-3906-4b40-ae4e-2fbca7fe01e0/beautiful-scenery-wallpapers-moonlight-beauty.jpg"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "image_captioning",
      image: "https://replicate.delivery/mgxm/13f7ddbb-3906-4b40-ae4e-2fbca7fe01e0/beautiful-scenery-wallpapers-moonlight-beauty.jpg"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "image_captioning",
        "image": "https://replicate.delivery/mgxm/13f7ddbb-3906-4b40-ae4e-2fbca7fe01e0/beautiful-scenery-wallpapers-moonlight-beauty.jpg"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "image_captioning",
      "image": "https://replicate.delivery/mgxm/13f7ddbb-3906-4b40-ae4e-2fbca7fe01e0/beautiful-scenery-wallpapers-moonlight-beauty.jpg"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="image_captioning"' \
  -i 'image="https://replicate.delivery/mgxm/13f7ddbb-3906-4b40-ae4e-2fbca7fe01e0/beautiful-scenery-wallpapers-moonlight-beauty.jpg"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "image_captioning",
      "image": "https://replicate.delivery/mgxm/13f7ddbb-3906-4b40-ae4e-2fbca7fe01e0/beautiful-scenery-wallpapers-moonlight-beauty.jpg"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Caption: a painting of a lake surrounded by trees and flowers

Generated in

5.2 seconds

salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099

Prediction

Model

ID

mojxr2cdczexzjxsvjuvbummqi

Status

Succeeded

Source

Web

Hardware

–

Total duration

1.2s

Created

over 3 years ago

Input

task: visual_question_answering
image
question: what is standing on the rock?

{
  "task": "visual_question_answering",
  "image": "https://replicate.delivery/mgxm/ceba852d-1e52-4004-ae8b-86869790dbc3/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
  "question": "what is standing on the rock?"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "visual_question_answering",
      image: "https://replicate.delivery/mgxm/ceba852d-1e52-4004-ae8b-86869790dbc3/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      question: "what is standing on the rock?"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "visual_question_answering",
        "image": "https://replicate.delivery/mgxm/ceba852d-1e52-4004-ae8b-86869790dbc3/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
        "question": "what is standing on the rock?"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/ceba852d-1e52-4004-ae8b-86869790dbc3/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      "question": "what is standing on the rock?"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="visual_question_answering"' \
  -i 'image="https://replicate.delivery/mgxm/ceba852d-1e52-4004-ae8b-86869790dbc3/beautiful-scenery-wallpapers-moonlight-beauty.jpg"' \
  -i 'question="what is standing on the rock?"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/ceba852d-1e52-4004-ae8b-86869790dbc3/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      "question": "what is standing on the rock?"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Answer: deer

Generated in

0.9 seconds

salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099

Prediction

Model

ID

r4m2l3fcjzdk5pcaw3yxddi7su

Status

Succeeded

Source

Web

Hardware

–

Total duration

1.3s

Created

over 3 years ago by @lijunnan1992

Input

task: visual_question_answering
image
question: what is in the lake?

{
  "task": "visual_question_answering",
  "image": "https://replicate.delivery/mgxm/efe546ab-97e8-417c-9716-33a9ff8f7514/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
  "question": "what is in the lake?"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "visual_question_answering",
      image: "https://replicate.delivery/mgxm/efe546ab-97e8-417c-9716-33a9ff8f7514/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      question: "what is in the lake?"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "visual_question_answering",
        "image": "https://replicate.delivery/mgxm/efe546ab-97e8-417c-9716-33a9ff8f7514/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
        "question": "what is in the lake?"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/efe546ab-97e8-417c-9716-33a9ff8f7514/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      "question": "what is in the lake?"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="visual_question_answering"' \
  -i 'image="https://replicate.delivery/mgxm/efe546ab-97e8-417c-9716-33a9ff8f7514/beautiful-scenery-wallpapers-moonlight-beauty.jpg"' \
  -i 'question="what is in the lake?"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/efe546ab-97e8-417c-9716-33a9ff8f7514/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      "question": "what is in the lake?"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Answer: swans

Generated in

1.0 second

salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099

Prediction

Model

ID

hsru35hlhnd53ilh6o3fq4d7s4

Status

Succeeded

Source

Web

Hardware

–

Total duration

1.2s

Created

over 3 years ago

Input

task: visual_question_answering
image
question: what is in the sky?

{
  "task": "visual_question_answering",
  "image": "https://replicate.delivery/mgxm/7ce95df7-5794-42dc-bd3c-068ef8c12c2f/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
  "question": "what is in the sky?"
}

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
  {
    input: {
      task: "visual_question_answering",
      image: "https://replicate.delivery/mgxm/7ce95df7-5794-42dc-bd3c-068ef8c12c2f/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      question: "what is in the sky?"
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

pip install replicate

Import the client:

import replicate

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    input={
        "task": "visual_question_answering",
        "image": "https://replicate.delivery/mgxm/7ce95df7-5794-42dc-bd3c-068ef8c12c2f/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
        "question": "what is in the sky?"
    }
)

# The salesforce/blip model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/salesforce/blip/api#output-schema
    print(item)

To learn more, take a look at the guide on getting started with Python.

Run salesforce/blip using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "salesforce/blip:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099",
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/7ce95df7-5794-42dc-bd3c-068ef8c12c2f/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      "question": "what is in the sky?"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099 \
  -i 'task="visual_question_answering"' \
  -i 'image="https://replicate.delivery/mgxm/7ce95df7-5794-42dc-bd3c-068ef8c12c2f/beautiful-scenery-wallpapers-moonlight-beauty.jpg"' \
  -i 'question="what is in the sky?"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/salesforce/blip@sha256:5a977fcb091363cc3938cc870c9f6242f5938267040e6837373387bb90d23099
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "task": "visual_question_answering",
      "image": "https://replicate.delivery/mgxm/7ce95df7-5794-42dc-bd3c-068ef8c12c2f/beautiful-scenery-wallpapers-moonlight-beauty.jpg",
      "question": "what is in the sky?"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Answer: moon

Generated in

1.0 second